Abstract Book

S867

ESTRO 37

the start of Abiraterone was 15.0 months (95%CI=1.5- 37.2). Progression-free survival calculated from the end of radiotherapy was 5.2 months (95%CI=3.9-6.5). Conclusion Conclusion: Our series demonstrated that radiotherapy might prolong Abiraterone treatment in mCRPC patients. More patients and comparative studies are mandatory to verify this result and to evaluate survival improvement. EP-1611 Decision support for rectum spacers: combining tumor control, rectum toxicity and genetic markers Y. Van Wijk 1 , B. Vanneste 1 , A. Jochems 1 , S. Walsh 1 , P. Lambin 1 1 Maastricht university, GROW - School for Oncology and Developmental Biology, Maastricht, The Netherlands Purpose or Objective Dose escalated radiotherapy for prostate cancer patients has revealed higher tumor control probabilities (TCP) [Zelefsky et al. Eur Urol. 2011] but consequently increases the normal tissue complication probability (NTCP) in the rectum. An implantable rectum spacer (IRS) reduces the high dose in the rectum, and could allow for more dose escalation, without increasing toxicity. An iso- toxic model was developed to calculate the amount of dose escalation possible for a given treatment plan with a chosen threshold for the NTCP. A virtual IRS (V-IRS) [van Wijk et al. Radiother Oncol. 2017], which used image deformation to obain CT images with IRS before IRS placement, was applied. The iso-toxic model in combination with the V-IRS could be used to support the decision to place an IRS in a given patient. Material and Methods This Proof of Concept study included 16 localized prostate cancer patients with an IRS. Treatment planning was performed on computed tomography (CT) images before and after IRS placement and on images with a V- IRS. For these plans the normal tissue complication probability (NTCP) was calculated using a Lyman-Kutcher- Burman model for Grade>=2 late rectal bleeding [Lui et al. Acta Oncol. 2010], which was expanded with single- nucleotide polymorphisms (SNP) for late rectal toxicity [Kerns et al. EBioMedicine 2010]. The model adjusted the number of dose fractions, recalculating the equivalent dose volume histogram and resulting NTCP, until the NTCP was under 5%. The resulting treatment plan was used to calculate the TCP using a published model [Walsh et al. Med Phys. 2012]. Results Figure 1 shows the median TCP and NTCP plotted against the number of fractions for the treatment plans without IRS (solid line) and with IRS (dashed). The median TCP for patients without an IRS was 77.2% [39.2-91.4], and the placement of an IRS resulted in an increased median TCP of 96.3% [80.8-99.2]. The median TCP of the V-IRS was 96.3% [80.8-99.7], showing that the V-IRS has the potential to accurately predict the TCP resulting from a real IRS (Figure 2). If the same patients were heterogeneous for the SNP’s for rectal toxicity, the median TCP without IRS would be 48.9% [15.9-72.3], but with IRS this increases to 92.9% [61.6-99.1] without exceeding the 5% NTCP threshold.

Figure 1 - TCP: Tumor control probability; NTCP: Normal Tissue Complication Probability; IRS: implantable rectum spacer; nf: number of fractions

Figure 2 - IRS: implantable rectum spacer; TCP: tumor control probability; NTCP: normal tissue complication

probability Conclusion

In this study we showed that placing an IRS can potentially improve the outcome for prostate cancer patients when combined with dose escalation. Additionally, results suggested that the placement of an IRS could improve radiotherapy outcome in patients heterogeneous for SNP’s correlated with rectum toxicity. We concluded that the developed model together with a V-IRS has potential to serve as a multifactorial decision support system for the placement of an IRS. EP-1612 Pre-treatment mpMRI accuracy and prognostic value in intermediate risk localized prostate cancer D. Büchser 1 , A. Gómez-Iturriaga 1 , F. Casquero 1 , A. Urresola 2 , A. Ezquerro 2 , J. Cacicedo 1 , I. San Miguel 1 , F. Suarez 1 , P. Bilbao 1 1 Hospital Universitario Cruces, Radiation Oncology, Barakaldo, Spain 2 Hospital Universitario Cruces, Radiology, Barakaldo, Spain Purpose or Objective Intermediate-risk prostate cancer represents the largest of the risk groups and is comprised of a heterogeneous